The RealReal tightens bot defenses after AI agents exploited reserve system
The luxury consignment retailer updated its security following reports that Meta's Muse could indefinitely block other shoppers from purchasing items.
What happened
The RealReal announced it is updating its bot defense system to make it harder for AI agents to shop on its platform. The change follows reporting by The Verge that demonstrated Meta's Muse AI agent could be used to exploit The RealReal's 20-minute item reserve system—by continuously re-adding items to a shopping cart as holds expired, an agent could effectively block other shoppers from purchasing for hours or longer. The retailer's chief technology officer Luke Friang said the company has now made it "much more difficult" for agents to log in, add items to cart, and check out, though agents will still be able to surface listings and direct users to product pages.
Context
AI agents are emerging as a significant shopping use case for major tech platforms; Meta reported that shopping is the most common use case for Muse, with average purchases over $200, and that 10 million US users downloaded the app in 40 days. This creates a tension: while agents offer convenience as personalized shopping assistants, their ability to execute repetitive tasks at inhuman scale can break the assumptions underlying retail platforms—in this case, a simple inventory hold mechanism. Major retailers including Amazon have already begun blocking agents to protect their business models and customer experience. As more consumers gain access to shopping agents and learn to exploit platform features, retailers face an ongoing arms race in defensive measures.